<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=Windows-1252">
<style type="text/css" style="display:none;"> P {margin-top:0;margin-bottom:0;} </style>
</head>
<body dir="ltr">
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0); background-color: rgb(255, 255, 255);" class="elementToProof">
Hi everyone,</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0); background-color: rgb(255, 255, 255);" class="elementToProof">
<br>
</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0); background-color: rgb(255, 255, 255);" class="elementToProof">
In next week's CaCL meeting we will discuss Meister and Cotterell 2021.</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0); background-color: rgb(255, 255, 255);" class="elementToProof">
<br>
</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0); background-color: rgb(255, 255, 255);" class="elementToProof ContentPasted0">
<b>Title:</b> Language Model Evaluation Beyond Perplexity</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0); background-color: rgb(255, 255, 255);" class="elementToProof ContentPasted0">
<br>
</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0); background-color: rgb(255, 255, 255);" class="elementToProof ContentPasted0 ContentPasted2">
<b>Link:</b> <a href="https://aclanthology.org/2021.acl-long.414.pdf" id="LPlnk406651">https://aclanthology.org/2021.acl-long.414.pdf</a></div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0); background-color: rgb(255, 255, 255);" class="elementToProof ContentPasted0">
<br>
</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0); background-color: rgb(255, 255, 255);" class="elementToProof ContentPasted0">
<b>Abstract:</b></div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0); background-color: rgb(255, 255, 255);" class="elementToProof ContentPasted0">
<span class="ContentPasted1">We propose an alternate approach to quantifying how well language models learn natural language: we ask how well they match the statistical tendencies of natural language. To answer this question, we analyze whether text generated
from language models exhibits the statistical tendencies present in the human-generated text on which they were trained. We provide a framework–paired with significance tests–for evaluating the fit of language models to these trends. We find that neural language
models appear to learn only a subset of the tendencies considered, but align much more closely with empirical trends than proposed theoretical distributions (when present). Further, the fit to different distributions is highly-dependent on both model architecture
and generation strategy. As concrete examples, text generated under the nucleus sampling scheme adheres more closely to the type–token relationship of natural language than text produced using standard ancestral sampling; text from LSTMs reflects the natural
language distributions over length, stopwords, and symbols surprisingly well.<br>
</span></div>
<div class="elementToProof">
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
<br>
</div>
<div id="Signature">
<div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
<span style="font-size:12pt;margin:0px;text-align:start;background-color:rgb(255, 255, 255)"><span style="margin:0px">----</span></span>
<div style="font-size:12pt;margin:0px;text-align:start;background-color:rgb(255, 255, 255)">
</div>
<div style="font-size:12pt;margin:0px;text-align:start;background-color:rgb(255, 255, 255)">
<b>Christian Clark</b><br>
</div>
<div style="font-size:12pt;margin:0px;text-align:start;background-color:rgb(255, 255, 255)">
<div style="margin:0px">Ph.D. Student</div>
<div style="margin:0px">Department of Linguistics</div>
<span style="margin:0px">The Ohio State University</span></div>
</div>
</div>
</div>
</div>
</body>
</html>